The use of optical fibers for such analytic purposes is known and described, for example, in Hirschfeld. U.S. Pat. No. 4,577,109, entitled "Remote Multi-Position Information Gathering System and Method." Such techniques are also described by F. P. Milanovich, T. B. Hirschfeld, F. T. Wang, S. M. Klainer, and D. Walt in "Novel Optical Fiber Techniques for Medical Application," published in the Proceedings of the SPIE 28th Annual International Technical Symposium on Optics and Electrooptics. Volume 494.
Briefly summarizing the method of the prior art, a sensor consisting typically of a fluorescent dye is attached to the distal end of an optical fiber, preferably of diameter suitable for in vivo application. Light of a suitable wavelength, from an appropriate source, is used to illuminate the distal end of the optical fiber, visible light, for example, having a wavelength typically between about 400 and 800 nm. The light is propagated along the fiber to its distal end. A fraction of the light is reflected at the fiber-liquid interface and returns along the fiber. A second fraction of the light is absorbed by the fluorescent dye, and light is then emitted therefrom in a band centered usually but not always at a longer wavelength. The intensity of this band is dependent upon the interaction of the dye with the property or substance measured in the body fluid or tissue. This fluorescent light is also propagated along the return path of the fiber and measured using suitable optical equipment. The ratio of the wavelengths is determined and the parameter of interest measured.
Fluorescent molecules have been attached to the distal end of an optical fiber or to a separate piece of glass, e.g., a glass bead, by a technique which is used in the immobilization of enzymes. See Methods of Enzymology, vol. XLIV, Ed. Klaus Mosbach, Academic Press. pp. 134-148 (1976). A fundamental problem encountered with such a technique has been the failure to provide sufficient amounts of indicator at the distal end of the fiber. The prior art describes two primary approaches to the problem. In the first of these, a porous substrate is used so that more surface area is available to which indicator moieties may bind, while in the second, as described, for example, in U.S. Pat. No. 4,269,516 to Lubbers et al., indicator is provided in solution form, which solution is separated from the external environment by a membrane.
In the former procedure, a glass surface, for example the distal end of an optical fiber (or a glass bead which is subsequently attached to the fiber), is treated so that it is porous. It is then reacted with a suitable agent such as 3-aminopropyltriethoxy silane, and a series of reactions is carried out culminating in the covalent attachment of a biologically and/or chemically active molecule, e.g., a fluorescent species, to the surface of the glass. These reactions may be represented as follows: ##STR1## Now representing the silanized glass (referred to also hereinafter as aminoalkyl glass or fiber) as ##STR2## the next step is realized by reacting a fluorophore such as fluorescein isothiocyanate dissolved in an aprotic solvent with the aminoalkyl fiber ##STR3## and performing the remaining conventional washing and purification steps.
A dye immobilized on glass in this manner is pH-sensitive in the physiological range and is relatively stable. However, to achieve a satisfactory degree of response, it has been necessary to employ porous glass, that is, a special alkali borosilicate glass which has been heat treated to separate the glass into two phases. Typically, the phase which is rich in boric acid and alkali is leached out with acid to leave a porous, high-silica structure. The porous structure is then silanized and treated with the desired fluorescent species. The resulting treated glass is then attached to the distal end of an optical fiber. Porous glass provides more surface area available for silanizing and thus many more fluorescent molecules can be attached.
However, this technique is difficult to carry out. The most practical way to accomplish the intended result is to provide a bead of porous glass, silanize it, react it with the fluorescent species, and then attach the glass bead to the distal end of an optical fiber. This attachment is difficult to effect because of the small size of the beads and the ease with which the pores in the glass are occluded.
The technique of using a solution behind a membrane also suffers from serious drawbacks. Most importantly, the phenomena of concentration quenching comes into play, which severely limits the amount of dye that can be in close proximity to the end of the fiber. Where fluorescent dyes are present in concentrations higher than about 10.sup.-3 M, for example, concentration quenching occurs and results in a substantial loss of accuracy. Several processes are believed to be responsible for concentration quenching: (1) the increased probability of self-absorption at higher densities or concentrations of fluorophores; (2) formation of dimers or higher aggregates which are normally less fluorescent than the monomer; and (3) reaction of excited molecules with ground state molecules to form excimers, which again have emission spectra quite different from the monomer spectrum. See, e.g., Guilbault, Practical Fluorescence, New York: Marcel Dekker, Inc., 1973. Similar problems arise in the case of absorbing dyes, where exciton interaction--that is, interaction of dipoles on neighboring groups or molecules--is a factor at higher densities. See Birks, Photophysics of Aromatic Molecules, London: Wiley & Sons, 1970.